git/list[1] front-page[2] threads[3] people[4] search[5] about
 

[PATCH/RFC 1/2] send-pack --stateless-rpc: properly close the outgoing channel

From
Johannes Sixt <j6t@kdbg.org>
Date
Apr 24, 2011, 20:42 UTC
Message-ID
<4DB48B2C.2090904@kdbg.org>
In-Reply-To
<7vsjtkfs10.fsf@alter.siamese.dyndns.org>

In the non-stateless-rpc case, the writable end of the channel to the remote repo is closed by the start_command() call that runs the pack-objects process (after pack-objects inherited a copy). But in the --stateless-rpc case, where send-pack takes care of writing data to the channel, this was missed.

Signed-off-by: Johannes Sixt <j6t@kdbg.org>
---
Am 14.04.2011 23:21, schrieb Junio C Hamano:
Show 8 quoted lines
> Jeff King <peff@peff.net> writes:
> 
>> I think I am leaning a bit towards (2a). It's simple, and it's not like
>> this is library code with a million unknown callers; fixing it simply
>> and cleanly with a nice commit message is probably sufficient.
> 
> This really sounds like a plan.  Even if we _might_ later want to go to
> 2b. or some other solution, we will know what pattern to grep for.

Here's a 2-patch series that implements this plan. The patches go on top of 38a81b4e (receive-pack: Wrap status reports inside side-band-64k) just like Jeff's series (jk/maint-push-async-hang).

Warning: This patch is untested. Furthermore, it does not even fix a resource
leak because the fd that is now closed in pack_objects() would be closed
later in cmd_send_pack. However, without closing the fd earlier like this,
a --stateless-rpc invocation could theoretically dead-lock just like a regular
invocation in a NO_PTHREADS build. But I also don't know how to test-drive
send-pack --stateless-rpc to construct such a case. Any hints how to do that
would be appreciated.
-- Hannes
 builtin-send-pack.c |    1 +
 1 files changed, 1 insertions(+), 0 deletions(-)
diff --git a/builtin-send-pack.c b/builtin-send-pack.c
index 2478e18..089058b 100644
--- a/builtin-send-pack.c
+++ b/builtin-send-pack.c
@@ -97,6 +97,7 @@ static int pack_objects(int fd, struct ref *refs, struct extra_have_objects *ext
 		free(buf);
 		close(po.out);
 		po.out = -1;
+		close(fd);
 	}
 
 	if (finish_command(&po))
-- 
1.7.5.rc1.97.ge0653
Previous: Junio C HamanoNext: Johannes Sixt
Message 18 of 48 in “fix hang in git push when pack-objects fails”
  1. 0/4 fix hang in git push when pack-objects failsJeff King, Mar 31, 2011
  2. 1/4 teach wait_or_whine a "quiet" modeJeff King, Mar 31, 2011
  3. Johannes SixtMar 31, 2011
  4. Jeff KingApr 1, 2011
  5. 2/4 finish_async: be quiet when waiting for async processJeff King, Mar 31, 2011
  6. 3/4 run-command: allow aborting async code prematurelyJeff King, Mar 31, 2011
  7. Erik Faye-LundApr 1, 2011
  8. Jeff KingApr 1, 2011
  9. 4/4 send-pack: abort sideband demuxer on pack-objects errorJeff King, Mar 31, 2011
  10. Johannes SixtApr 13, 2011
  11. Jeff KingApr 14, 2011
  12. Johannes SixtApr 14, 2011
  13. Jeff KingApr 14, 2011
  14. Johannes SixtApr 14, 2011
  15. Jeff KingApr 14, 2011
  16. Johannes SixtApr 14, 2011
  17. Junio C HamanoApr 14, 2011
  18. 1/2 send-pack --stateless-rpc: properly close the outgoing channelJohannes Sixt, Apr 24, 2011
  19. 2/2 send-pack: avoid deadlock when pack-object dies earlyJohannes Sixt, Apr 24, 2011
  20. Jeff KingApr 25, 2011
  21. Johannes SixtApr 25, 2011
  22. Junio C HamanoApr 25, 2011
  23. send-pack: avoid deadlock when pack-object dies earlyJohannes Sixt, Apr 25, 2011
  24. Jeff KingApr 26, 2011
  25. Jeff KingApr 25, 2011
  26. 5/4 run-command: implement abort_async for pthreadsJeff King, Mar 31, 2011
  27. Erik Faye-LundApr 1, 2011
  28. Erik Faye-LundApr 1, 2011
  29. Johannes SixtApr 1, 2011
  30. Jeff KingApr 1, 2011
  31. Erik Faye-LundApr 1, 2011
  32. Erik Faye-LundApr 1, 2011
  33. Johannes SixtApr 1, 2011
  34. Erik Faye-LundApr 1, 2011
  35. Jeff KingApr 1, 2011
  36. Erik Faye-LundApr 1, 2011
  37. Jeff KingApr 1, 2011
  38. Jeff KingApr 1, 2011
  39. Erik Faye-LundApr 1, 2011
  40. Johannes SixtApr 1, 2011
  41. Erik Faye-LundApr 1, 2011
  42. Johannes SixtApr 1, 2011
  43. Erik Faye-LundApr 1, 2011
  44. Jeff KingApr 1, 2011
  45. Erik Faye-LundApr 2, 2011
  46. Jeff KingApr 1, 2011
  47. Johannes SixtMar 31, 2011
  48. Jeff KingApr 1, 2011

Read the whole thread, see it on lore, or plain text.

$ cat FOOTERMessages come from the public archive at lore.kernel.org/git, fetched every hour. The front page is chosen and written each morning by an AI editor and can be wrong; the threads themselves are the record. About and API. For agents: an MCP server at https://gitlist.dev/mcp, and any thread, story or person page as Markdown by adding .md to its URL (or sending Accept: text/markdown). Details in /llms.txt.